The American jobs market is facing a significant downturn, as a preliminary report from the Bureau of Labor Statistics (BLS) unveiled a staggering adjustment to job growth figures. The report cited that the US economy added approximately 911,000 fewer jobs than initially estimated for the year ending in March. Should this preliminary adjustment hold, it would mark the most substantial annual revision to US employment data on record, with the final annual benchmark expected to be reported in February 2026.
This release represents the first phase of the BLS’s annual benchmark review, a practice that has evolved for nearly 90 years. Each year, the BLS endeavors to provide a more accurate employment count by reconciling data from business surveys—which are timely yet less precise—with the more accurate but significantly delayed unemployment insurance quarterly tax filings.
The preliminary benchmark revision of -911,000 exceeds economists’ predictions and accounts for about 0.6% of total employment. Over the past decade, annual benchmark revisions have averaged a mere 0.2% of total nonfarm employment, indicating the magnitude of this year’s unexpected adjustment.
If the downward revision is confirmed, the average monthly job growth for the period from April 2024 to March 2025 would decline significantly, from the current figure of 146,500 positions per month to roughly 70,500. This projected reduction approximates a nearly 76,000 drop in monthly job gains.
Economists have posited that this massive downward revision is largely attributed to pandemic-related disruptions affecting the so-called birth-death model, a statistical tool designed to estimate business and job creation. Prior assessments had anticipated a downturn due to several factors: less-than-expected job creation at new firms, sampling errors stemming from decreasing response rates for surveys, and necessary adjustments for asylum-seekers and undocumented workers.
According to Pantheon Macroeconomics, a substantial portion of the revision appears to stem from the birth-death model inaccuracies, with estimates suggesting that two-thirds of the adjustment may be related to weaker job creation at new businesses than previously inferred.
Sector analysis reveals that one-quarter of the preliminary estimated revision occurred within the trade, transportation, and utilities sector, resulting in a downward adjustment of 226,000 jobs, or 0.8% of employment. Conversely, the information sector, which includes technology firms, is set to experience the most significant percentage-based revision, with a preliminary estimate indicating a loss of 67,000 jobs, accounting for 2.3%.
It is crucial to understand that federal data is subject to frequent revisions as more accurate information becomes available. The BLS’s monthly jobs report is designed to provide timely insights into employment trends, but this immediacy comes at the expense of precision.
To compile monthly payroll estimates, the BLS surveys around 120,000 US employers, representing about 600,000 work sites—approximately one-third of total employment. Employers are granted multiple opportunities to report their payroll gains and losses each month.
The annual revision process involves adjusting the data collected from monthly payroll surveys and benchmarking the employment figures from March against those derived from the Quarterly Census of Employment and Wages (QCEW) program. This program provides a comprehensive assessment of businesses, employees, and wages at various geographical levels, utilizing quarterly tax data submitted by companies, although this method is inherently lagged.
As the situation develops, further updates will be provided to keep the public informed about this significant adjustment in the American job market.
